Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/19517
標題: 人型機器人之模仿能力探討與實作
A imitation system for humanoid robot with implementation
作者: 張人祐
Jhang, Ren-You
關鍵字: neural network
人型機器人
humanoid robot
self-learning
motion imitation
動作模仿
自我學習
類神經網路
出版社: 資訊科學與工程學系所
引用: [1]葉怡成,類神經網路模式應用與實作, 儒林圖書,2003. [2]范逸之、江文賢、陳立元, C++ Builder 與 RS-232串列通訊, 文魁資訊, 2004. [3]洪靜宜,Visual C+ + .NET 深入研究 第六版. [4]黃文吉,C++ Builder與影像處理, 儒林圖書,2005. [5]游天龍,”彩色影像辨識技術應用於機器人運動控之研究 “,全國碩博士論文網,2005.06。 [6]周立柏,”具模仿能力之智慧型雙足機器人之設計與實現”,全國碩博士論文網 ,2007. 06。 [7]偉克多微電腦工作室,http://vic.seeder.net/ [8]WowWee 娛樂機器人公司, http://www.wowwee.com/ [9]HONDA ASIMO機器人, http://world.honda.com/ASIMO/ [10] Atkeson,C.G. and Schaal,S.” Robot Learning From Demonstration.” Proceedings of the Fourteenth International Conference (ICML 『97)、Edited by Douglas H. Fisher,Jr. pp. 12-20、Morgan Kaufmann、San Francisco,CA,1997. [11] Demiris,J. and Hayes,G..” Imitative Learning Mechanisms in Robots and Humans.“ Proceedings of the 5th European Workshop on Learning Robots,July 1996. [12] Demiris,J.andMataric,M.J.,“ Perceptuo-Motor Primitives in Imitation .” Workshop on Agents in Interaction-Acquiring Competence Through Imitation,2nd International Conference on Autonomous Agents(Agents 『98),Minneapolis/St. Paul,May 10-13,1998. [13] Williams,T. G. , Rowland,J. J. , Lee,M.H. and Neal,M.J.” Teaching by Example in Food Assembly by Robot.” Proceedings of the 2000 IEEE, International Conference on Robotics & Automation. Vol.4, 24-28 April 2000 PP:3247 – 3252. [14] Maja J. Mataric′,“Getting Humanoids to Move and Imitate.” IEEE,Intelligent systems and their application.Vol.15,NO.4,2000,pp 18-24. [15] Xiaoyuan He, Ryo Kojima and Osamu Hasegawa.” Developmental Word Grounding Through a Growing Neural Network With a Humanoid Robot.” IEEE Trans, Syst., Man, Cybern., Vol. 37,NO.2,2007,pp451-462. [16] F. Shen and O. Hasegawa, “An incremental network for on-line unsupervised classificationand topology learning,” Neural Netw., vol. 19, no. 1, pp. 90–106, Jan. 2006. [17] D. Roy, K. Hsiao, and N. Mavridis, “Mental imagery for a conversational robot,” IEEE Trans. Syst., Man, Cybern. B, Cybern., vol. 34, no. 3, pp. 1374–1383, Jun. 2004. [18] S. Schaal, C.G. Atkeson, and S.Vijayakumar,“Real-Time Robot Learning with Locally Weighted Statistical Learning,” Proc. IEEE Int’l Conf. Robotics and Automation, IEEE Press, Piscataway, N.J., 2000, pp. 288–293. [19] J. Elman, “Learning and development in neural networks: The importance of starting small.” Cognition. vol. 48, no. 1, pp. 71–99, 1993. [20] Xiaoyuan He, Kojima, R.; Hasegawa, O. ”Developmental Word Grounding Through a Growing Neural Network With a Humanoid Robot.” IEEE Trans. Syst., Man, Cybern.Vol 37, Issue 2, April.2007. [21] Li-Po Chou, Wen-June Wang,” A Humanoid Robot with Motion Imitation Ability” International Conference on Machine Learning and Cybernetics .Vol4, pp.2031-2036, 19-22 Aug. 2007. [22] Okada, M.; Nakamura and Y.; Hoshino, S.” Development of the cybernetic shoulder-a three DOF mechanism that imitates biological shoulder-motion.” Proceedings of the 2000 IEEE,International Conference on Intelligent Robots and Systems. Vol. 1, 17-21 Oct. 1999 PP:543 - 548.
摘要: 在兩足機器人領域中,由於多自由度與動作複雜度的設計門檻高,完善的動作模仿系統可以有效的改善並且加速機器人開發效率。本論文主要為設計一個具有學習能力的動作模仿系統,並在兩足機器人平台上實作之。在設計系統方面,採用了較特殊的想法,主要想把每個動作都能簡化成類似漫畫的線條,再加以學習模仿,如此可達成不需要任何第三者定位、開發者不需計算任何有關動力學的數學、不需配戴任何感應器或易於辨識的東西以及不需使用的昂貴的感應器(如超音波、雷射等)以上特色。也由於利用了類神經網路的特性,系統可以記憶學習過的動作,並且在接受新的動作時不需要讓開發者重新設計程式,如應用在生活中,將可大幅降低開發費用。實驗中,採用了現成的兩足機器人以及簡易架設的視訊平台,僅靠平面影像辨識做為動作分析的輸入資訊,將所獲得的動作圖像以新的演算法將之簡化成動作線條是本論文最大的貢獻。
The humanoid robot has become more and more popular. Account of the high devise threshold of multi degree of freedom and complicated motions, a thorough motion imitation system enables us to improve and accelerate the effectiveness of the research and development of robot. There are many methods about motion imitation, like being located by a third party, calculating mathematics about dynamics, wearing sensors or something is easily identified, and using costly sensors(such as ultrasound or laser). But this research is mainly to design a motion imitation system with learning ability and approve it on the humanoid robot. Adopting some kind of new ideas in the aspect of devise system, all motions are to be simplified to lines like cartoons. Besides, learning and imitate, we are proposing a new imitative robot system based on Back-propagation neural network (BNN). The objective of this work is to implement imitative system based on humanoid robot that can imitate human's motions and can recall all motions that have been trained.
URI: http://hdl.handle.net/11455/19517
其他識別: U0005-0907200813440800
文章連結: http://www.airitilibrary.com/Publication/alDetailedMesh1?DocID=U0005-0907200813440800
Appears in Collections:資訊科學與工程學系所

文件中的檔案:

取得全文請前往華藝線上圖書館



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.